Artificial intelligence and machine learning could help in the development of functional food

Since the trends of healthy eating, health and environmental care, as well as zero waste living and sustainability have entered consumers’ lives, an increasing number of buyers are questioning where their food comes from, what ingredients it contains, how many resources were used for it, and where it goes.

The food industry has long been aware of this and directs its production, distribution, and marketing activities towards these trends, but in this entire process, advanced technology, in the form of ubiquitous artificial intelligence and digitalization, also offers new practical solutions.

AI for functionality

Given the growth, but still insufficiently high awareness of the importance of healthy eating among people in Croatia, the fifteenth expert meeting ‘Functional Food in Croatia’ was recently held, organized by the Croatian Chamber of Commerce and the Croatian Society for Food Technology, Biotechnology, and Nutrition. Due to the growing market and increasing demand for food that offers health benefits, the opportunity for the food industry lies in the production of functional food, emphasized Jelena Đugum, director of the Agriculture Sector of the Croatian Chamber of Commerce.

– Increasing investment in research in this area is crucial for improving innovation and competitiveness. Collaboration with research institutions, the scientific community, and other companies to jointly finance such projects should be a strong driver of growth in the functional and innovative food market in Croatia – Đugum added.

The benefits of machine learning

To make the process of developing functional food simpler, faster, and more efficient, one of the solutions is the application of artificial intelligence and machine learning. As stated by Prof. Dr. Nenad Bolf from the Faculty of Chemical Engineering and Technology at the University of Zagreb in his contribution published in the proceedings ‘Application in Research and Production of Functional Food’, algorithms can analyze vast datasets of ‘chemical and biological information to identify potential bioactive compounds in food’.

‘These algorithms can predict therapeutic properties, which is crucial for designing functional food. For example, researchers can use machine learning to identify the optimal combination of ingredients for creating an energy bar with a high antioxidant content or probiotic yogurt that supports gut health. This not only reduces research time but also minimizes waste in the product development phase,’ Bolf states in the proceedings.

In this way, companies can produce functional food with the help of technology for increasingly demanding consumers, whose benefits, in addition to basic functions, manifest in health benefits, such as reducing the risk of certain chronic diseases. The term ‘functional food’ originated in Japan, the country with the longest-lived population in the world, back in the 1980s. As reported by the Croatian Chamber of Commerce, by the end of 2022, the global market for functional food reached a value of 193.77 billion US dollars and is expected to continue growing at an annual rate of 6.7 percent, thus the potential for producers in this segment is significant.

Production and distribution traceability

Due to increasing indicators of the negative impact of food on our health, the issue of food traceability is gaining importance today, emphasizes Matija Žulj, CEO and founder of Agrivi, a company that deals with the digitalization of agriculture. As Žulj notes, according to the UN, more than 600 million cases of foodborne illnesses are recorded each year worldwide, and there are significantly more cases that are not recorded.

– When we talk about food traceability, most producers and consumers focus on the origin of the product, where the food came from, and what path it took from the field to the shelf. It is important to realize that even more important information is how that food was produced because production practices determine, for example, how many pesticides were used in crop production or how many antibiotics in animal husbandry. Thus, we can divide traceability into production traceability and distribution traceability.

For production traceability, there are software solutions for managing plant and animal production that provide insight into all essential information about the production method, while for distribution traceability, solutions that record the path and conditions at the location through which a product passes are most commonly applied. For example, just as you can track where a postal shipment is at any given moment – Žulj explains.

Agrivi is currently developing a software platform in Croatia for managing the agricultural supply chain, i.e., for the traceability of all agricultural processes. It is intended for the food and beverage industry that directly purchases agricultural products and provides information about every stage of the production path of a food product – the origin of the crops with information about the farm and field from which the product comes, the way the product was produced, what it was treated with, information about the nutritional values of the product, and so on.

– The key advantage of this technology is that it provides consumers with transparency and trust in the food they consume, as they can track the origin of the product and the processes it has undergone. On the other hand, it enables farmers, food and beverage producers, and retail chains to manage their operations more efficiently, improve traceability and transparency in the supply chain, facilitating the tracking of their products and meeting regulatory requirements and compliance with food safety regulations – Žulj notes.

Concrete results

That technology can be key in saving a company from unwanted costs, even bankruptcy, is evidenced by the example of Agrivi’s client who produces food for children, which is a particularly sensitive category where it is essential to ensure maximum quality and safety of food.

– For the production of fruit and vegetable purees, the client contracts the production of apples, carrots, and other fruits and vegetables with farmers. The situation where, for example, excess pesticides are found in children’s puree can lead the company to bankruptcy due to high costs of withdrawing products from the market and consumer reactions that will lose trust in the products of that company – Žulj emphasizes.

Technology, he believes, can significantly contribute to resource savings throughout the food supply chain, and he supports this with concrete examples.

– From our experience working with farmers around the world, we see that users of our software record concrete results such as, for example, a 15 percent optimization in water use for irrigation, a 13 percent reduction in fertilizer use, an eight percent reduction in fuel consumption in the fields, and a 21 percent reduction in pesticide use. These are truly tangible results when we talk about the impact of food on our health, as we are what we eat, and the reduction of the negative impact of agricultural production on the environment, especially if we know that the agricultural-food industry is one of the larger emitters of greenhouse gases – Žulj concludes.